- Email: [email protected]
Improved harness crutch to reduce upper limb effort in swing-through gait
Papers
Improved harness cru :ch to reduce upper limb effort in swing-t xrough gait B. J. Andrews, M.H. Granat, B. W. Heller, J. MacMahon”, S. Real*
L. Keating* and
Bioengineering Unit, Wolfson Centre, University of Strathclyde, 106 Rottenrow, Glasgow G4 ONW, Scotland and *School of Physiotherapy, Mater Misericardiane Hospital, Eccles Street, Dublin 7, Ireland ABSTRACT A novel crutch, the harness crutch, is &scribed which reduces loading on the arms during the swing phase of swingthrough gait. l?te &ice wa.s fabricated by attaching a modified mountainem’ng harness by two side straps, to mo@ed axiL2zry crutches. 7’7~ harness crutch was compared with the saddle crutch, dtxribed by Taylor in 1883. 17re sadaYe in excess of 500 m&g in ti perirwal area; no fiessures were produced in thtk area with th-e crutch producedpressures harnwc cmtch Ischial pressures produced by ‘both systems were similar. In sk out of eight non-impaired subjects, sign$cantiy morefice ~4s transmit&d to the harness crutch (an average 47% of body weight) than to the saddle crutch (an average 40% of btiy weigh). A comparison of the oxygnr cost of swing-through gait was ma& between the harness crutch and unmod$ed axillq crutches; with the hamess crutch oxygen cost was sign@antly lower (p < 0.01) and there were no signijcant difknces in speed and stria2 length.
Keywords: Swing-throughgait, paraplegicgait, crutches,oxygen cost
INTRODUCTION Crutches have been used to assist gait for nearly 5000 years’ and yet their basic design has not significantly changed in that time2. For patients who use crutches, swing-through gait is often preferred as it is fast when compared with other forms of crutch-aided gait. Subjects who perform swing-through gait prefer axillary crutches, which are still routinely used3 owin to their subjective impressions of increased stabili 2 . A high level of upper-limb mu&&u activity is shoulder joint % similar to thole at the hi in unaided gait6. The up er limbs are not suiteB for weight bearing as loa & are transferred across the shoulder joint by muscle action in comparison with the passive load transfer (by the interaction of joint surfaces) in the lower limbs. The high loads on the upper limbs lead to rapid muscular fatigue which limits maximum walking distance. Prolonged use of the upper limbs for wei ht bearing may lead to degeneration of the shoul ! er joint7. Other problems associated with regular high limb loading which paraplegics, who use their upper limbs for transfer and wheelchair propulsion, are prone to are chronic impingement syndrome and tearing of the rotator cuff *. Bayley’ has suggested that this may be caused by high inter-articular pressures. In addition to the problems involving the shoulder Correspondence and reprint requests to: Dr M.H. Granat 0 1994 Butterworth-Heinemann for BES 1350-4533/94/01015-04
joint, repeated hand activity has been related to c al tunnel synclromeg. % e mechanical energy expenditure of swing‘t is roughly equivalent to that of normal metabolic costs are much higher. suggested that this is because mechanical work is being done by the upper limbs rather than the legs, a view supported in a study by McBeath” who found that non-wei t-bearing crutch gait required 78% more energy $ an normal gait, whilst partialweight-bearing crutch gait required 33% more energy. To reduce upper limb loading, Taylor12 proposed the use of a saddle crutch in which a saddle is suspended between two axillary crutches. The saddle transfers force to the crutches via the perineal area during the swing phase of the gait. He reported satisfactory result5 on several fracture patients; the authors know of no reports that quantify the benefits of the saddle crutch. The system has the potential disadvantage of producing high pressures in the sensitive perineum area. The motivation for the present study is the application of functional electrical stimulation to assist paraplegics to perform swingthrough gait13’14. The aims of this study were to:
(1)develop a walking aid which transfers forces
away from the arms during the swing phase of gait., but which does not produce high pressures in the perineum area. (2) Compare this system with the saddle crutch. Med. Eng. Phys. 1994, Vol. 16, January
15
Improvedhamm crutch:BJ Andrcwset al.
which gave a resolution of 0.91 N. The sampling frequency was 50Hz. Interface pressures between the subject and the walking aids were recorded using a dynamic interface pressure monitoring system15. Pressure transducers were placed under both ischia. An additional transducer was laced in the perineum area for the saddle crutch trialp s. Gait trials Eight non-im aired subjects (four male and four female aged ! O-37) were selected for these trials. Each subject was familiarized with the two devices by trained physiotherapists.
Figure 1 The harness crutch. During the swing phase of gait, forces are transferred to the straps which connect the harness to the modified crutches. The buckles, between the thigh straps and tbe waist belt and those between the waist belt and the crutch, allow the adjustment of the system
(3) Compare the metabolic requirements of the system with those of standard axillary crutches.
METHOD System development A saddle crutch was constructed using the methods of Taylor12. The saddle was made from leather with attachments of standard grade mountaineering webbing. A harness crutch was made from a climbing harness, modified to allow loads to be taken at the sides, rather than at the front. The harness crutch consists of a waist band and two thigh bands which transmit the forces from the body to the crutches via two adjustable side straps (Rere 7). The two side straps of the saddle or harness were attached to aluminium rings (karabiners) which were connected to hooks attached to modified axillary crutches below the axilla pad. The lengths of the side straps were adjusted for each subject to optimize ground clearance during the swing phase. Both karabiners were strain gauged using four strain gauges (Type EA-06-062TT-120, Micromeasurement, Michigan, USA) mounted in a Wheatstone bridge arrangement on one arm of each karabiner. This permitted the measurement of axial loading only. Forces transmitted through the karabiners were recorded on a IBM-compatible PC using a 12-bit analogue-to-digital (A/D) converter,
16
Med. Eng. Phys. 1994, Vol. 16,. January
Comparison of the saddle crutch and the harness crutch. The subjects performed five straight-line 20m trials over a flat, level walkway with each system. Only data collected durin the middle 15m of the trial were analysed. In ad j ition to monitoring loadings, the time and number of steps required to complete 15 m were recorded. The parameters calculated for these trials were: average speed; average step length; and average maximum force transmitted through the side stra s in each step (as a percentage of the patient’s bo By weight). Additionally, in one set of trials, pressures under the ischia and perineum were recorded. Comparison of the harness crutch and unmodajied crutches. The oxygen consumption of subjects performing swing-through gait using the harness crutch and using standard crutches was compared. Oxygen cost was calculated by subtracting the resting oxygen consumption from that when performing swing-through gait and dividing the result by the distance travelled. Douglas bags were used to collect the expired air and an oximeter measured the percentage of oxygen in the ex ired air. During gait, the Douglas bags were CarrieB by an investigator walking alongside the subject. Each subject erformed each gait, around a large continuous waI! path, for a period of five minutes. Expired air was collected for the last two minutes of the gait. Between the tests each subject rested until his/her heart rate returned to the normal resting value. Pressure(mmHg)
2.00
4.00
6.00
Time Figure 2
Time course of perineum pressures for the saddle crutch
Improvedhamess crutch.~ BJ Andwws et al.
Percentage
Pressure (mmHg)
body weight
r
6
8
10
12
14
Time
C
Figure 3
Time course of ischial pressures for the saddle crutch and the harness crutch
Percentage body weight
1...._..........._....._.... SW _....._........_._
60
20
. . .. . . . . . ..
. . . . . . _.._.
. .._ . . . . . . . . . ..-
-..-..-.---.
.--...
_ __.......
-._ . .._.__._
G
H
SuD4jectsE Figure 5 Percentage weight bearing of 0 the saddle crutch and ? ? the harness crutch for each subject. Means and standard deviations are shown for the five gait trial each subject performed with each system
Oxygen cost (mVkg/m) 0.5
__.__
. . . .
Time Figure 4
Loading cycle of the harness crutch
Subjects RESULTS A ical time course of perineum pressure for the sadd ‘ype crutch, during swing-through gait, (Figure 2) shows the high pressures that the perineum area is regularly subjected to using this s tern. The magnitudes of the ischial pressures an c? their time course (F&ure 3), for the saddle crutch and the harness crutch, are similar. The loading cycle of the harness crutch system (Figure 4) during part of a gait trial shows the consistency of the loading profile. The system is loaded to over 50% of the subject’s body weight during the swing phase of the gait, thus providing a substantial reduction in the forces that the upper limbs must transmit. F&ure 5 shows the maximum force, expressed as a percentage of body weight, transmitted through both the saddle crutch and harness crutch for all eight subjects. In six subjects, the harness crutch took more force (p < 0.05, ANOVA). There were no significant differences in speed and stride length between the saddle crutch and the harness crutch. Seven subjects showed a reduction in oxygen cost and one an increase, when corn aring the harness crutch with unmodified crutches PFigure 6). Oxy en costs for the harness crutch were significantly Hess than for unmodified crutches (p < 0.0 1, two-tailed
Oxygen costs of ? ? the harness crutch and 0 unmodified crutches. One gait trial for each subject using each system was made. Oxygen costs were calculated as oxygen consumption (in ml per kilogram of body weight per metre distance travelled) during the last two minutes of the five-minute gait trial
Figure 6
Table 1 Comparison between unmodified axillary crutches and the harness crutch. Means and standard deviations (given in parentheses) are calculated for the eight subjects Unmodified crutches
Harness crutch
02 consumption (ml kg-’ m-‘)
0.34 (0.07)
0.25 (0.08)
Speed (m SK’)
0.85 (0.14)
0.83 (0.15)
Stride length (m)
1.45 (0.14)
1.41 (0.13)
Wilcoxon matched-pairs signed-rank test). There was no significant difference in speed and stride length between the harness crutch and the unmodified crutches (Table 7). DISCUSSION High pressures in the perineum (in excess of 500mmHg) in the saddle crutch trials made the system uncomfortable for use by sensate subjects. In patients where there is sensory deficit (for instance.
Med. Eng. Phys. 1994, Vol. 16, January
17
ImpodhamessmutchcBJ An&us et al.
those with spinal cord injuries) it may be unsuitable owing to the relationship between high pressures and the incidence of pressure sores16. An alternative harness crutch system has been developed in which force is transferred via the upper parts of the thighs and the hips. In this system there were no pressures on the perineum and the ischial pressures were similar to those of the saddle crutch. Subjects found the harness crutch to be more comfortable. The average percentage force transmitted to the harness crutch was 47% of body weight. This represents a si ificant reduction in upper-limb loading and there Yr ore provides potential benefit for patients usin swing-through gait. In most subjects, significan f y more force was taken by the harness crutch than the saddle crutch. Even in sub’ects where more force is taken by the saddle crutcll the use of the harness crutch may be preferable owing to the absence of erineal pressure. Overall, tKe harness crutch gave a 27% reduction in oxygen cost without any significant changes in speed or stride length. This may reflect the reduction in the use of the upper limbs for weight bearing during the swing phase of gait. We have characterized various aspects of the saddle crutch and develo ed a harness crutch which has several advantages. s ur immediate application is for FES-assisted swing-through gait14 in which the knees are unlocked and are actively flexed during the swing phase. Upper-limb effort is one of the major limitations of this gait. Most subjects preferred the harness crutch, however some reported an impression of lower stability than for unmodified crutches. We propose to address this problem by looking at design modifications. The reductions in upper limb loading and metabolic cost make this system an important advance in crutch design. In addition, the side stra s prevent incorrect use of the axillary crutches an B thus may prevent axillary artery thrombosis’6. This system would be applicable to any patient population performing swing-through gait with free knees. Further trials of this system on various patient populations are now indicated. CONCLUSIONS Our studies have shown that use of either the harness crutch or the saddle crutch can significantly reduce swing phase loadings of the upper limbs in normals performing swing-through gait with free knees. The saddle crutch produces potentially harmful pressures on the perineum area whereas the harness crutch does not. The oxygen cost of swing-through gait is signifi-
18 Med. Eng. Phys. 1994, Vol. 16,Jammy
cantly reduced when using the harness compared with unmodified crutches.
crutch
ACKNOWLEDGEMENTS The financial support of the Scottish Home and Health Department and the Science and Engineering Research Council (SERC) is gratefully acknowledged. -CES Epstein S. Art, history and crutch. Ann Med History, 1937; 9: 304-13. Shoup TE, Fletcher LS, Merril BR. Biomechanics of crutch locomotion. JBiomech 1974; 7: 11-19. Hall J, Clarke AK. An evaluation of crutches. Physiotlurapy 1991; 77: 156-60. Sankarankutty M, Stallard J, Rose GK. The relative efficiency of ‘swing-through’ gait on axillary elbow and Canadian crutches compared to normal walking. JBiomed Eng 1979; 1: 55-7. 5. Peacock B. A myographic and photographic study of walking with crutches. Physiotherapr London 1966; 52:
264-8. 6. Opila ISA, Nicol AC, Paul JP. Upper limb loadings of gait with crutches. J Biomech Eng 1987; 109: 285-90. 7. Gellman H, Sie I, Waters RL. Late complications of the weight-bearing upper extremity in the paraplegic patient. Clin Orthop 1988; 233: 132-5. 8. Bayley JC, Co&ran TP, Sledge CB. The weight bearing shoulder: the impingement syndrome in paraplegics.
JBone Joint Surg 1987; 69A: 182. 9. Aljure MD, Eltorai I, Bradley WE, Lin JE, Johnson B. Carpal tunnel syndrome in paraplegic patients. Paraplegia
1985; 23: 182-6. 10. Wells RP. The kinematics and energy variations of swingthrough crutch gait. JBionech 1979; 12: 579-85. 11. McBeath AA, Bahrke M, Balke B. Efficiency of assisted ambulation determined by oxygen consumption measurement. JBoneJoint Surg [Am] 1974; 56: 994-1000. 12. Taylor JR. A new saddle crutch. 2% Medical Record 1883;
24: 136. 13. Heller BW, Andrews BJ. An analysis of swinging gaits and their synthesis using functional electrical stimulation. Proc
3rd ViennaInternational Work.&@on Functional Ekxtrostimu&ion, Baden-Badm, Vienna, Sept 1989, 77-80. 14. Heller BW, Granat MH, Kirkwood CA, Delargy M. Preliminary studies of swing throught gait using FES.
Advances in Extemal Control of Human Extremities 1990; X: 225-32. 15. Barbanel JC, Sockahngham S. Device for measuring soft tissue interface pressures. J Biomed Eng 1990; 12: 5 19-22. 16. Kosiak M. Etiology of decubitus ulcers. Arch Phys Med 1961; 42: 19-29. 17. Brooks AL, Fowler SB. Axillary artery thrombosis after prolonged use of crutches. JBone Joint Surg DrnJ 1964; 46: 863-4.
Improved harness cru :ch to reduce upper limb effort in swing-t xrough gait B. J. Andrews, M.H. Granat, B. W. Heller, J. MacMahon”, S. Real*
L. Keating* and
Bioengineering Unit, Wolfson Centre, University of Strathclyde, 106 Rottenrow, Glasgow G4 ONW, Scotland and *School of Physiotherapy, Mater Misericardiane Hospital, Eccles Street, Dublin 7, Ireland ABSTRACT A novel crutch, the harness crutch, is &scribed which reduces loading on the arms during the swing phase of swingthrough gait. l?te &ice wa.s fabricated by attaching a modified mountainem’ng harness by two side straps, to mo@ed axiL2zry crutches. 7’7~ harness crutch was compared with the saddle crutch, dtxribed by Taylor in 1883. 17re sadaYe in excess of 500 m&g in ti perirwal area; no fiessures were produced in thtk area with th-e crutch producedpressures harnwc cmtch Ischial pressures produced by ‘both systems were similar. In sk out of eight non-impaired subjects, sign$cantiy morefice ~4s transmit&d to the harness crutch (an average 47% of body weight) than to the saddle crutch (an average 40% of btiy weigh). A comparison of the oxygnr cost of swing-through gait was ma& between the harness crutch and unmod$ed axillq crutches; with the hamess crutch oxygen cost was sign@antly lower (p < 0.01) and there were no signijcant difknces in speed and stria2 length.
Keywords: Swing-throughgait, paraplegicgait, crutches,oxygen cost
INTRODUCTION Crutches have been used to assist gait for nearly 5000 years’ and yet their basic design has not significantly changed in that time2. For patients who use crutches, swing-through gait is often preferred as it is fast when compared with other forms of crutch-aided gait. Subjects who perform swing-through gait prefer axillary crutches, which are still routinely used3 owin to their subjective impressions of increased stabili 2 . A high level of upper-limb mu&&u activity is shoulder joint % similar to thole at the hi in unaided gait6. The up er limbs are not suiteB for weight bearing as loa & are transferred across the shoulder joint by muscle action in comparison with the passive load transfer (by the interaction of joint surfaces) in the lower limbs. The high loads on the upper limbs lead to rapid muscular fatigue which limits maximum walking distance. Prolonged use of the upper limbs for wei ht bearing may lead to degeneration of the shoul ! er joint7. Other problems associated with regular high limb loading which paraplegics, who use their upper limbs for transfer and wheelchair propulsion, are prone to are chronic impingement syndrome and tearing of the rotator cuff *. Bayley’ has suggested that this may be caused by high inter-articular pressures. In addition to the problems involving the shoulder Correspondence and reprint requests to: Dr M.H. Granat 0 1994 Butterworth-Heinemann for BES 1350-4533/94/01015-04
joint, repeated hand activity has been related to c al tunnel synclromeg. % e mechanical energy expenditure of swing‘t is roughly equivalent to that of normal metabolic costs are much higher. suggested that this is because mechanical work is being done by the upper limbs rather than the legs, a view supported in a study by McBeath” who found that non-wei t-bearing crutch gait required 78% more energy $ an normal gait, whilst partialweight-bearing crutch gait required 33% more energy. To reduce upper limb loading, Taylor12 proposed the use of a saddle crutch in which a saddle is suspended between two axillary crutches. The saddle transfers force to the crutches via the perineal area during the swing phase of the gait. He reported satisfactory result5 on several fracture patients; the authors know of no reports that quantify the benefits of the saddle crutch. The system has the potential disadvantage of producing high pressures in the sensitive perineum area. The motivation for the present study is the application of functional electrical stimulation to assist paraplegics to perform swingthrough gait13’14. The aims of this study were to:
(1)develop a walking aid which transfers forces
away from the arms during the swing phase of gait., but which does not produce high pressures in the perineum area. (2) Compare this system with the saddle crutch. Med. Eng. Phys. 1994, Vol. 16, January
15
Improvedhamm crutch:BJ Andrcwset al.
which gave a resolution of 0.91 N. The sampling frequency was 50Hz. Interface pressures between the subject and the walking aids were recorded using a dynamic interface pressure monitoring system15. Pressure transducers were placed under both ischia. An additional transducer was laced in the perineum area for the saddle crutch trialp s. Gait trials Eight non-im aired subjects (four male and four female aged ! O-37) were selected for these trials. Each subject was familiarized with the two devices by trained physiotherapists.
Figure 1 The harness crutch. During the swing phase of gait, forces are transferred to the straps which connect the harness to the modified crutches. The buckles, between the thigh straps and tbe waist belt and those between the waist belt and the crutch, allow the adjustment of the system
(3) Compare the metabolic requirements of the system with those of standard axillary crutches.
METHOD System development A saddle crutch was constructed using the methods of Taylor12. The saddle was made from leather with attachments of standard grade mountaineering webbing. A harness crutch was made from a climbing harness, modified to allow loads to be taken at the sides, rather than at the front. The harness crutch consists of a waist band and two thigh bands which transmit the forces from the body to the crutches via two adjustable side straps (Rere 7). The two side straps of the saddle or harness were attached to aluminium rings (karabiners) which were connected to hooks attached to modified axillary crutches below the axilla pad. The lengths of the side straps were adjusted for each subject to optimize ground clearance during the swing phase. Both karabiners were strain gauged using four strain gauges (Type EA-06-062TT-120, Micromeasurement, Michigan, USA) mounted in a Wheatstone bridge arrangement on one arm of each karabiner. This permitted the measurement of axial loading only. Forces transmitted through the karabiners were recorded on a IBM-compatible PC using a 12-bit analogue-to-digital (A/D) converter,
16
Med. Eng. Phys. 1994, Vol. 16,. January
Comparison of the saddle crutch and the harness crutch. The subjects performed five straight-line 20m trials over a flat, level walkway with each system. Only data collected durin the middle 15m of the trial were analysed. In ad j ition to monitoring loadings, the time and number of steps required to complete 15 m were recorded. The parameters calculated for these trials were: average speed; average step length; and average maximum force transmitted through the side stra s in each step (as a percentage of the patient’s bo By weight). Additionally, in one set of trials, pressures under the ischia and perineum were recorded. Comparison of the harness crutch and unmodajied crutches. The oxygen consumption of subjects performing swing-through gait using the harness crutch and using standard crutches was compared. Oxygen cost was calculated by subtracting the resting oxygen consumption from that when performing swing-through gait and dividing the result by the distance travelled. Douglas bags were used to collect the expired air and an oximeter measured the percentage of oxygen in the ex ired air. During gait, the Douglas bags were CarrieB by an investigator walking alongside the subject. Each subject erformed each gait, around a large continuous waI! path, for a period of five minutes. Expired air was collected for the last two minutes of the gait. Between the tests each subject rested until his/her heart rate returned to the normal resting value. Pressure(mmHg)
2.00
4.00
6.00
Time Figure 2
Time course of perineum pressures for the saddle crutch
Improvedhamess crutch.~ BJ Andwws et al.
Percentage
Pressure (mmHg)
body weight
r
6
8
10
12
14
Time
C
Figure 3
Time course of ischial pressures for the saddle crutch and the harness crutch
Percentage body weight
1...._..........._....._.... SW _....._........_._
60
20
. . .. . . . . . ..
. . . . . . _.._.
. .._ . . . . . . . . . ..-
-..-..-.---.
.--...
_ __.......
-._ . .._.__._
G
H
SuD4jectsE Figure 5 Percentage weight bearing of 0 the saddle crutch and ? ? the harness crutch for each subject. Means and standard deviations are shown for the five gait trial each subject performed with each system
Oxygen cost (mVkg/m) 0.5
__.__
. . . .
Time Figure 4
Loading cycle of the harness crutch
Subjects RESULTS A ical time course of perineum pressure for the sadd ‘ype crutch, during swing-through gait, (Figure 2) shows the high pressures that the perineum area is regularly subjected to using this s tern. The magnitudes of the ischial pressures an c? their time course (F&ure 3), for the saddle crutch and the harness crutch, are similar. The loading cycle of the harness crutch system (Figure 4) during part of a gait trial shows the consistency of the loading profile. The system is loaded to over 50% of the subject’s body weight during the swing phase of the gait, thus providing a substantial reduction in the forces that the upper limbs must transmit. F&ure 5 shows the maximum force, expressed as a percentage of body weight, transmitted through both the saddle crutch and harness crutch for all eight subjects. In six subjects, the harness crutch took more force (p < 0.05, ANOVA). There were no significant differences in speed and stride length between the saddle crutch and the harness crutch. Seven subjects showed a reduction in oxygen cost and one an increase, when corn aring the harness crutch with unmodified crutches PFigure 6). Oxy en costs for the harness crutch were significantly Hess than for unmodified crutches (p < 0.0 1, two-tailed
Oxygen costs of ? ? the harness crutch and 0 unmodified crutches. One gait trial for each subject using each system was made. Oxygen costs were calculated as oxygen consumption (in ml per kilogram of body weight per metre distance travelled) during the last two minutes of the five-minute gait trial
Figure 6
Table 1 Comparison between unmodified axillary crutches and the harness crutch. Means and standard deviations (given in parentheses) are calculated for the eight subjects Unmodified crutches
Harness crutch
02 consumption (ml kg-’ m-‘)
0.34 (0.07)
0.25 (0.08)
Speed (m SK’)
0.85 (0.14)
0.83 (0.15)
Stride length (m)
1.45 (0.14)
1.41 (0.13)
Wilcoxon matched-pairs signed-rank test). There was no significant difference in speed and stride length between the harness crutch and the unmodified crutches (Table 7). DISCUSSION High pressures in the perineum (in excess of 500mmHg) in the saddle crutch trials made the system uncomfortable for use by sensate subjects. In patients where there is sensory deficit (for instance.
Med. Eng. Phys. 1994, Vol. 16, January
17
ImpodhamessmutchcBJ An&us et al.
those with spinal cord injuries) it may be unsuitable owing to the relationship between high pressures and the incidence of pressure sores16. An alternative harness crutch system has been developed in which force is transferred via the upper parts of the thighs and the hips. In this system there were no pressures on the perineum and the ischial pressures were similar to those of the saddle crutch. Subjects found the harness crutch to be more comfortable. The average percentage force transmitted to the harness crutch was 47% of body weight. This represents a si ificant reduction in upper-limb loading and there Yr ore provides potential benefit for patients usin swing-through gait. In most subjects, significan f y more force was taken by the harness crutch than the saddle crutch. Even in sub’ects where more force is taken by the saddle crutcll the use of the harness crutch may be preferable owing to the absence of erineal pressure. Overall, tKe harness crutch gave a 27% reduction in oxygen cost without any significant changes in speed or stride length. This may reflect the reduction in the use of the upper limbs for weight bearing during the swing phase of gait. We have characterized various aspects of the saddle crutch and develo ed a harness crutch which has several advantages. s ur immediate application is for FES-assisted swing-through gait14 in which the knees are unlocked and are actively flexed during the swing phase. Upper-limb effort is one of the major limitations of this gait. Most subjects preferred the harness crutch, however some reported an impression of lower stability than for unmodified crutches. We propose to address this problem by looking at design modifications. The reductions in upper limb loading and metabolic cost make this system an important advance in crutch design. In addition, the side stra s prevent incorrect use of the axillary crutches an B thus may prevent axillary artery thrombosis’6. This system would be applicable to any patient population performing swing-through gait with free knees. Further trials of this system on various patient populations are now indicated. CONCLUSIONS Our studies have shown that use of either the harness crutch or the saddle crutch can significantly reduce swing phase loadings of the upper limbs in normals performing swing-through gait with free knees. The saddle crutch produces potentially harmful pressures on the perineum area whereas the harness crutch does not. The oxygen cost of swing-through gait is signifi-
18 Med. Eng. Phys. 1994, Vol. 16,Jammy
cantly reduced when using the harness compared with unmodified crutches.
crutch
ACKNOWLEDGEMENTS The financial support of the Scottish Home and Health Department and the Science and Engineering Research Council (SERC) is gratefully acknowledged. -CES Epstein S. Art, history and crutch. Ann Med History, 1937; 9: 304-13. Shoup TE, Fletcher LS, Merril BR. Biomechanics of crutch locomotion. JBiomech 1974; 7: 11-19. Hall J, Clarke AK. An evaluation of crutches. Physiotlurapy 1991; 77: 156-60. Sankarankutty M, Stallard J, Rose GK. The relative efficiency of ‘swing-through’ gait on axillary elbow and Canadian crutches compared to normal walking. JBiomed Eng 1979; 1: 55-7. 5. Peacock B. A myographic and photographic study of walking with crutches. Physiotherapr London 1966; 52:
264-8. 6. Opila ISA, Nicol AC, Paul JP. Upper limb loadings of gait with crutches. J Biomech Eng 1987; 109: 285-90. 7. Gellman H, Sie I, Waters RL. Late complications of the weight-bearing upper extremity in the paraplegic patient. Clin Orthop 1988; 233: 132-5. 8. Bayley JC, Co&ran TP, Sledge CB. The weight bearing shoulder: the impingement syndrome in paraplegics.
JBone Joint Surg 1987; 69A: 182. 9. Aljure MD, Eltorai I, Bradley WE, Lin JE, Johnson B. Carpal tunnel syndrome in paraplegic patients. Paraplegia
1985; 23: 182-6. 10. Wells RP. The kinematics and energy variations of swingthrough crutch gait. JBionech 1979; 12: 579-85. 11. McBeath AA, Bahrke M, Balke B. Efficiency of assisted ambulation determined by oxygen consumption measurement. JBoneJoint Surg [Am] 1974; 56: 994-1000. 12. Taylor JR. A new saddle crutch. 2% Medical Record 1883;
24: 136. 13. Heller BW, Andrews BJ. An analysis of swinging gaits and their synthesis using functional electrical stimulation. Proc
3rd ViennaInternational Work.&@on Functional Ekxtrostimu&ion, Baden-Badm, Vienna, Sept 1989, 77-80. 14. Heller BW, Granat MH, Kirkwood CA, Delargy M. Preliminary studies of swing throught gait using FES.
Advances in Extemal Control of Human Extremities 1990; X: 225-32. 15. Barbanel JC, Sockahngham S. Device for measuring soft tissue interface pressures. J Biomed Eng 1990; 12: 5 19-22. 16. Kosiak M. Etiology of decubitus ulcers. Arch Phys Med 1961; 42: 19-29. 17. Brooks AL, Fowler SB. Axillary artery thrombosis after prolonged use of crutches. JBone Joint Surg DrnJ 1964; 46: 863-4.